|
|
Post by WP 257 on Apr 14, 2015 10:33:37 GMT -8
Intermountain non-sound GEVOs come with a unique decoder. The purpose of the decoder is so that when they are paired with a DCC/Sound equipped unit, they will run the same speed as the full blown sound-equipped unit, even when operating in plain DC mode.
Intermountain did this to save some folks money by allowing them to mix sound and non-sound units in the same lashup, even in plain DC. It was a novel solution.
Due to higher than typical starting voltage, and slower top speed in plain DC mode (at least at 12 volts), their decision has been criticized by many people. Intermountain's GEVOs can be run (are designed to be run) at voltages higher than 12 volts, which has also angered some people who don't want to buy certain power supplies that will go up to 16 or even 18 volts in plain DC.
They are not the only manufacturer to have produced engines that can be safely operated (without frying the motor) at voltages above 12 in plain DC. Some earlier BLI engines could run on 16 or even 18 volts, and have a built in function that totally shuts down the unit when the voltage gets too high, to avoid damage.
|
|
|
|
Post by llxlocomotives on Apr 14, 2015 11:11:00 GMT -8
DC has been 16 volts for more than twenty years. All, the MEC tech II supplies I've tested measure slightly more than 16 volts. Most DC motors are set up to run safely at 18 volts. This is one of the problems I have with DCC, you only have 10 volts of potential when you have 14 available in the unit (assuming it starts at 2 volts). That effects load as well as speed.
|
|
|
|
Post by mlehman on Apr 14, 2015 23:44:29 GMT -8
SNIP Most DC motors are set up to run safely at 18 volts. This is one of the problems I have with DCC, you only have 10 volts of potential when you have 14 available in the unit (assuming it starts at 2 volts). That effects load as well as speed. Not quite sure abiout your numbers on DCC. Most systems comes set to operate around 14.5 volts. They're user-adjustable. My NCE PowerPro manual says 9.5 to 18 volts range. Mine are dialed down to 12.5 volts to baby my Micro-Tsunamis. Speaking of Tsunamis, in DC mode or Analog Mode as they call it, many of the features programmable in DCC are available in DC. So you can set starting volts, the speed table (not sure on this one), momentum effects, various sound features. Not everything makes the crossover. Aren't other sound decoders similar in offering many of the DCC features on DC? I rarely use mine on DC, but it's a nice option. |
|
|
|
Post by llxlocomotives on Apr 15, 2015 4:38:12 GMT -8
I was referring to the maximum motor voltage. The ones I have measured indicate just under 12 volts when my NCE test unit is set at the max speed step.
Because I rehab and repair all type of HO engines, I find this type data important.
My experience with sound engines, which is not compete, indicates that on DC they follow a specific program. A start up sound program and then the motor starts. This is keyed to the external DC voltage level. I haven't had my hands on any of the latest popular sound systems. The ones I see are not very adjustable in DC.
|
|
|
|
Post by mlehman on Apr 15, 2015 7:15:53 GMT -8
Considering most locomotives on most layouts spend about 0.03% of total operating hours at top speed -- unless your name is Gomez Addams -- I guess it's something I never knew I was missing  I suspect motor output from the decoder to the motor is governed by the NMRA specs, which is why is just happens to be so similar -- it's intended that way. And it's true that most DC motors can run at a higher voltage than nominal specs indicate. Nothing new about that. But if you're an engineer, someone has to pic a number so that things written into manufacturing specs one place can be replicated half a world away. But the Industrial Revolution sort of made the idea of standardization uncontroversial a couple of centuries ago, I thought. Yes, any time you run supply voltage through a device, you're likely to have some voltage drop. Lots of reason for that. Good example is a bridge rectifier. And loads cause voltage drop, take your ubiquitous wallwart. If you measured the voltage on the back side of a DC motor, I suspect you'll see voltage drop there, too. I guess I see this as an issue in search of being a problem, because I just don't see how it makes any difference most model railroaders would ever notice. Now, if you're writing the script for Furious 8 -- FastTracks, Revenge of the Blue-Box, maybe you're onto something... |
|
|
|
Post by MONSTERRAILROAD on Apr 15, 2015 7:52:24 GMT -8
Considering most locomotives on most layouts spend about 0.03% of total operating hours at top speed -- unless your name is Gomez Addams -- I guess it's something I never knew I was missing I suspect motor output from the decoder to the motor is governed by the NMRA specs, which is why is just happens to be so similar -- it's intended that way. And it's true that most DC motors can run at a higher voltage than nominal specs indicate. Nothing new about that. But if you're an engineer, someone has to pic a number so that things written into manufacturing specs one place can be replicated half a world away. But the Industrial Revolution sort of made the idea of standardization uncontroversial a couple of centuries ago, I thought. Yes, any time you run supply voltage through a device, you're likely to have some voltage drop. Lots of reason for that. Good example is a bridge rectifier. And loads cause voltage drop, take your ubiquitous wallwart. If you measured the voltage on the back side of a DC motor, I suspect you'll see voltage drop there, too. I guess I see this as an issue in search of being a problem, because I just don't see how it makes any difference most model railroaders would ever notice. Now, if you're writing the script for Furious 8 -- FastTracks, Revenge of the Blue-Box, maybe you're onto something... LOL. Call me Gomez Addams filming Fast n Furious 8. I love Hi-balling! |
|
|
|
Post by llxlocomotives on Apr 15, 2015 8:09:03 GMT -8
At then end of the day it is power, which is voltage time current.
If you pulling a heavy train with resistance, you will need the power. To say it another way, a hundred mile per hour engine only capability will approach zero mph when pulling a train of 50-60 cars at NMRA weight. It only matters if you have grades long trains and sharp curves.
As a retired engineer, it bothers me to leave 30 percent of the power capability on the table.
|
|
|
|
Post by Mark R. on Apr 15, 2015 13:22:23 GMT -8
I was referring to the maximum motor voltage. The ones I have measured indicate just under 12 volts when my NCE test unit is set at the max speed step. Because I rehab and repair all type of HO engines, I find this type data important. My experience with sound engines, which is not compete, indicates that on DC they follow a specific program. A start up sound program and then the motor starts. This is keyed to the external DC voltage level. I haven't had my hands on any of the latest popular sound systems. The ones I see are not very adjustable in DC. What you are not understanding is that a decoder doesn't apply a variable voltage to the motor - it applies pulses (pulse width modulation) of the full voltage potential. This cannot be measured accurately using a typical meter. Mark. |
|
|
|
Post by llxlocomotives on Apr 15, 2015 16:24:38 GMT -8
I use a RRampmeter for my DCC measurements.
While not a sine curve, a pulse wave form is a voltage variation. As such it leads to less efficient motor performance.
I am not sure all decoders use PMW at all speed steps. Definitely at voltages below where the motor will run , but at power levels above that, the pulse does not provide any benefit. It just exasperates the motor heat issues.
|
|
|
|
Post by Mark R. on Apr 15, 2015 17:50:31 GMT -8
I use a RRampmeter for my DCC measurements. While not a sine curve, a pulse wave form is a voltage variation. As such it leads to less efficient motor performance. I am not sure all decoders use PMW at all speed steps. Definitely at voltages below where the motor will run , but at power levels above that, the pulse does not provide any benefit. It just exasperates the motor heat issues. You're confusing old school DC pulse power with DCC pulse width modulation. Unlike DC throttles, the motor is driven with a programmable waveform known as Pulse-Width Modulation (PWM), in which the maximum voltage is applied to the motor for some percentage of the time (versus DC, in which a percentage of the max voltage is applied all the time). For example, if we want the loco stopped, we apply max voltage 0% of the time (in DC, it's simply zero volts); if we want the loco to creep, DCC applies max voltage 10% of the time (in DC, we apply 10% of max voltage); if we want the loco to move at approx. half of top speed, DCC applies max voltage 50% of the time (in DC, it's 50% of max voltage all the time); for top speed, DCC applies max voltage all the time (so does DC). Confused? Figure 7 shows the DCC motor control waveforms graphically (the red line is the voltage actually applied to the loco's motor).  Mark. |
|
|
|
Post by llxlocomotives on Apr 15, 2015 18:19:30 GMT -8
You have described the theory very well. I not confused dr all. The realities are that one cycle occurs 10000 times a second. While very rapid, it is still a transient variation of viltage. It will be less efficient than a steady input. That inefficiency is heat, plain and simple.
The benefit for doing this is at low speed/power. Everywhere else it is a debit to the system. The reason the decoder manufacturer would want to do it this way would be to simplify the electronics.
|
|
|
|
Post by mlehman on Apr 15, 2015 18:43:00 GMT -8
You have described the theory very well. I not confused dr all. The realities are that one cycle occurs 10000 times a second. While very rapid, it is still a transient variation of viltage. It will be less efficient than a steady input. That inefficiency is heat, plain and simple. The benefit for doing this is at low speed/power. Everywhere else it is a debit to the system. The reason the decoder manufacturer would want to do it this way would be to simplify the electronics. I suspect it's true that the greatest comparative efficiencies would be at low speed, high loading for DC. There's a lot more given off as heat from a decoder in that state. On the other hand, that also means that the two efficiency curves converge at higher speeds and lighter loads. At anything other than close to full slip, the difference is so small as to be functionally irrelevant...but I'm not an engineer, so will let someone else handle the sliderule...  If we were talking about 1:1, then this would be worth arguing over in Engineering. But you'll never see the difference on your power bill in whatever smaller scale you run (assuming HO here since that's where thois started), so once you get away from overheating decoders, most model railroaders quickly decide they just don't have a dog in this fight. It's an interesting bit of trivia, but if you're running your stock that hard, the guys in Maintenance may be getting worried about how the whole system is going to hold up. Where's the weak link. because that's the results here if pushed to the bitter end. That's why guys blow up 1000+HP fuelies at the dragstrip every weekend -- and then are happy to have it back together for next Friday night. On the railroad, we call it adding a unit if you really are pushing things that hard. And if you're not, my guess is a quick and dirty on the back of a napkin will make this argument more of a cursiousity than a convention. 100,000s of DCC users seems to do OK. I don't think the historical precedents are in your favor here. Go back to the 1890s and listen to Edison and Tesla argue over DC vs AC for RR electrification. Sure, some DC seemed to work, but modern installs are pretty much AC. Why? Once you figure in transmission efficiencies and costs, whatever advantages DC has are washed away in the transmission costs. So if you want to scale the real world down, don't forget to take account of efficiencies to get the juice to you house, then convert it to DC, etc. Cherrypicking the data between the powerpack and the rails just doesn't tell the whole story here. |
|
|
|
Post by Mark R. on Apr 15, 2015 18:57:19 GMT -8
You have described the theory very well. I not confused dr all. The realities are that one cycle occurs 10000 times a second. While very rapid, it is still a transient variation of viltage. It will be less efficient than a steady input. That inefficiency is heat, plain and simple. The benefit for doing this is at low speed/power. Everywhere else it is a debit to the system. The reason the decoder manufacturer would want to do it this way would be to simplify the electronics. Actually, while being slightly noisier, this type of regulation will run much cooler than its linear counterpart. The use of pulse width modulation to control a motor has the advantage in that the power loss in the switching transistor is small because the transistor is either fully “ON” or fully “OFF”. As a result the switching transistor has a much reduced power dissipation giving it a linear type of control which results in better speed stability. Also the amplitude of the motor voltage remains constant so the motor is always at full strength. The result is that the motor can be rotated much more slowly without it stalling. Pulse width modulation is a great method of controlling the amount of power delivered to a motor without dissipating any wasted power (heat). It's not a matter of "simplfying the electronics" - it's a smarter more efficient way of controlling the electronics. Mark. |
|
